SEARCH

SEARCH BY CITATION

Keywords:

  • salamander;
  • limb development;
  • regeneration;
  • fossils;
  • branchiosaurids;
  • Lissamphibia

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. DEVELOPMENT
  5. REGENERATION
  6. WHAT THE FOSSILS TELL US
  7. DISCUSSION
  8. Acknowledgements
  9. REFERENCES

The development of the tetrapod limb during skeletogenesis follows a highly conservative pattern characterized by a general proximo-distal progression in the establishment of skeletal elements and a postaxial polarity in digit development. Salamanders represent the only exception to this pattern and display an early establishment of distal autopodial structures, specifically the basale commune, an amalgamation of distal carpal and tarsal 1 and 2, and a distinct preaxial polarity in digit development. This deviance from the conserved tetrapod pattern has resulted in a number of hypotheses to explain its developmental basis and evolutionary history. Here we summarize the current knowledge of salamander limb development under consideration of the fossil record to provide a deep time perspective of this evolutionary pathway and highlight what data will be needed in the future to gain a better understanding of salamander limb development specifically and tetrapod limb development and evolution more broadly. Developmental Dynamics 240:1087–1099, 2011. © 2011 Wiley-Liss, Inc.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. DEVELOPMENT
  5. REGENERATION
  6. WHAT THE FOSSILS TELL US
  7. DISCUSSION
  8. Acknowledgements
  9. REFERENCES

The limb is the hallmark of the tetrapod body plan and an extensive array of studies has focused on its development, morphology, and evolution (Hall, 2007). This organ system exemplifies how approaches from different fields of biology can be synthesized to yield more information than any one sub-discipline could provide. Despite such a research effort, many questions surrounding the development, evolution, and morphological diversification of the tetrapod limb remain unresolved (e.g., Shubin, 2002; Shubin et al., 2009; Zeller et al., 2009; Woltering and Duboule, 2010).

Salamanders play a central role in the discussion of tetrapod limb evolution because their skeletal development proceeds in a different way than other limbed tetrapods, i.e., anurans and amniotes. The differences refer to the patterning of skeletal elements, which follows a specific progression during the early condensation and chrondrification of mesenchymal cells that give rise to the individual segments of the limb. Anatomists recognized these patterns as early as the late 19th century, which sparked discussions on the early evolution and origin(s) of the tetrapod limb. Historically, Gegenbaur 1864 was the first to state that the salamander limb is the most ancestral of all tetrapods and derived it from a uniserial fin comparable to that of Neoceratodus, a concept that had been repeatedly revisited and discussed during the early days of research on tetrapod limb evolution (Holmgren, 1933, 1939, 1942; Jarvik, 1965). Strasser (1879) was among the first to provide a comprehensive description of the skeletal development in salamanders on the basis of the genera Salamandra, Triturus, Ichthyosaura, and Lissotriton (the latter three all termed Triton therein), which was followed by further detailed studies by a number of authors on urodeles as well as anurans and amniotes that set the framework for larger scale comparisons between different tetrapod clades (e.g., Wiedersheim, 1879; Zwick, 1898; Howes and Swinnerton, 1900; Pée, 1904; Schmalhausen, 1907, 1910, 1917; Steiner, 1921, 1934; Erdmann, 1933; Keller, 1946). Although many of the concepts of tetrapod limb evolution have changed over the course of the past century, these initial studies continue to provide invaluable data for our understanding of salamander limb development and evolution.

With the publication of Gould's Ontogeny and Phylogeny (Gould, 1977), the discovery of conserved factors underlying the evolution of vertebrate body plans and organs, and the subsequent rise of the field of evolutionary developmental biology, tetrapod limb development once more became a research focus, drawing attention from molecular biology, morphology, and paleontology alike. Salamanders continue to play an important role, albeit with a current focus on their regenerative capacities that allow them to regenerate whole or partial limbs as adult animals. Shubin and Alberch (1986) and Oster et al. (1988) revisited these issues elaborating major differences in timing and pattern of condensation between salamanders and other crown tetrapods (Fig. 1).

thumbnail image

Figure 1. Schematic representation of the patterns of limb skeletogenesis in amniotes and frogs (left) and salamanders (right). [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]

Download figure to PowerPoint

In amniotes and anurans, skeletogenesis generally proceeds in a proximodistal direction and commences with the condensation of the stylopodium (humerus/femur). The zeugopodial elements arise as a y-shaped condensation at the distal end of the humerus/femur with the preaxial element usually lagging behind the postaxial one. A number of further condensations establish the proximal elements of the carpus and tarsus. Finally, the distal carpals/tarsals and associated digits are formed, starting with the penultimate, postaxial digit (digit IV in a pentadactyl autopodium) and progressing preaxially in the order IV-(V)-III-II-I. Digit V often arises somewhat independently of the digital arch, although a connection to the remainder of the digital arch exists in some taxa. This characteristic pattern of postaxial dominance has been observed in all of the studied anurans and amniotes with surprisingly little variation in the basic progression of events, given the great morphological and functional diversity of tetrapod limbs.

Limb skeletogenesis of salamanders differs in a number of profound aspects from this conservative tetrapod pattern (Fig. 1). After the condensation of the stylopodium (humerus/femur), a y-shaped condensation proceeds to the zeugopodium, but contrary to other tetrapods, the preaxial element (radius/tibia) leads in development before the postaxial element (ulna/fibula). A further difference is an early and independent condensation of distal mesopodial elements before more proximal ones are established. This is reflected in an early appearance of the basale commune, an amalgamation of distal carpal/tarsal 1 and 2, which is known only from salamanders (Shubin and Wake, 2003) and the fossil amphibamid Gerobatrachus from the Early Permian of Texas (Anderson et al., 2008). Further carpals/tarsals form from mesenchymal extensions of more proximal elements, while the formation of the digital arch proceeds in a postaxial direction with an order of digit formation of II-I-III-IV(-V). Contrary to all other tetrapods, urodeles therefore show a distinct preaxial dominance in their pattern of limb skeletogenesis. Moreover, this reversed polarity is not only present during limb development, but has also been demonstrated for the developmental and evolutionary loss of digits in salamanders (Alberch and Gale, 1985; Stopper and Wagner, 2007).

Despite this striking deviance from an otherwise conserved pattern in tetrapods and the long history of research on salamander limb development, it remains unclear what the evolutionary history of this pattern is, what functional and/or phylogenetic significance it may have, and what role many of the classic limb development genes and their molecular control play in the formation of the salamander autopodium. It is essential to draw upon all available lines of evidence to gain a coherent picture of the evolution of salamander limb development specifically and by extension tetrapod limb development more generally.

DEVELOPMENT

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. DEVELOPMENT
  5. REGENERATION
  6. WHAT THE FOSSILS TELL US
  7. DISCUSSION
  8. Acknowledgements
  9. REFERENCES

A number of features in urodele limb development are very stable and can be observed in all limbed salamanders. These include the early development of the radius/tibia relative to the corresponding postaxial zeugopodial element, the early establishment of a basale commune in the fore- and hind limb, and the preaxial to postaxial differentiation of digits in the manus and pes (Shubin and Alberch, 1986; Shubin and Wake, 2003).

Other aspects of urodele limb development are more variable and reveal regional patterns of variation, i.e., variations in the connectivities and extension of mesenchymal condensation in the preaxial, central, and postaxial columns, and the digital arch, as well as ecological variablilty. Variation within urodeles appears to exceed that seen in other clades of tetrapod (Shubin and Alberch, 1986; Blanco and Alberch, 1992; Vorobyeva and Hinchliffe, 1996; Shubin and Wake, 2003; Franssen et al., 2005; Fröbisch, 2008).

One of the main sources of ontogenetic diversity lies in the overall timing of limb and digit formation, which varies between salamander taxa in accordance to their larval ecologies. Salamander taxa with pond larvae show a pronounced difference in timing between the fore- and hind limb, with the forelimb developing days in advance of the hind limb. This lag is less pronounced in salamanders that develop as larvae dwelling in streams, and even less so in direct developing salamanders (Blanco and Alberch, 1992; Wake and Shubin, 1998; Vorobyeva et al., 2000; Shubin and Wake, 2003). Differences in the timing of forelimb and hind limb development are known from many tetrapod clades (Bininda-Emonds et al., 2007), but they are particularly pronounced in urodeles.

Striking variation is also seen in the timing of differentiation of salamander digits. Instead of progressing through the characteristic paddle stage present in frogs and amniotes, salamanders with free-swimming aquatic larvae bud their digits one by one while the larvae are already swimming and interacting with their environment (Fig. 2A). The development of the early developing digits (II and I) is well advanced compared to subsequently forming digits. Digit formation in direct developing salamanders is more similar to amniotes and frogs in having a paddle stage and a more simultaneous development of the digits in the autopod, albeit still with a clear preaxial polarity (Franssen et al., 2005) (Fig. 2B).

thumbnail image

Figure 2. Series of forelimb skeletal development in the pond-larva of Ambystoma mexicanum (A) and the direct developing salamander Desmognathus aenus (B) illustrating the different pattern in digit formation. B is modified from Franssen et al. (2005).

Download figure to PowerPoint

Key signaling centers that control the differentiation of skeletal elements, such as the apical ectodermal ridge (AER) and zone of polarizing activity (ZPA), seem to be present in salamanders. However, although the conserved aspects of tetrapod limb development are often highlighted, there are differences that have been recognized. One prominent difference is the absence of a morphologically distinct apical ectodermal ridge in salamander limb buds (Tank et al., 1977). In tetrapods, a thickened cap of pseudostratified epithelium expressing a number of markers, including Fgfs, forms a feedback loop with shh in the ZPA, driving proliferation and outgrowth of the limb bud. In salamanders, the epidermis has AER-like properties and seems to fulfill the functions of the AER of other tetrapods but lacks any distinct thickened cap of tissue (Bryant and Muneoka, 1986; Torok et al., 1999b).

Major genes involved in limb patterning have been identified in salamanders, often with the purpose to compare their roles in the development and regeneration of the limb. Imokawa and Yoshizato (1997) and Torok et al. (1999b) showed that shh is expressed in the posterior region of the developing limb bud of axolotls and demonstrated with classic grafting experiments that it fulfills the ZPA polarizing activity as in developing amniote and frog limbs, determining the antero-posterior axis of the limb. Shh is expressed in the early limb bud stage, is strongest in mid limb bud stage and progressively weakened and shifted distally in the late limb bud stage (Imokawa and Yoshizato, 1997). Although the shh-expressing domain seems to be slightly smaller and more proximal in comparison to amniotes and frogs, possibly in accordance with the absence of an AER, the expression pattern is essentially comparable, indicating that salamander limbs have a ZPA that functions similarly to that of other tetrapods (Imokawa and Yoshizato, 1997; Torok et al., 1999a; Stopper and Wagner, 2007). Interestingly, Stopper and Wagner (2007) detected shh expression very late in limb development at the three-digit stage via an mRNA preparation of a proximal portion of the limb. The function of this late expression remains unclear, although based on the observations they made in inhibition experiments with cyclopamine, they state that it seems unlikely that shh functioned in the specification of anterior-posterior digit number in the drawn-out limb development of Ambystoma (Stopper and Wagner, 2007).

The expression pattern of Fgf-8 in the developing limb of salamanders also differs from that in other tetrapods, although expression elsewhere in the developing body, such as in facial primordia, brain, and tail bud, is highly conserved (Han et al., 2001; Christensen et al., 2002). In amniotes and frogs, Fgf-8 and Fgf-10 are expressed in the AER with a defined expression limit at the ectodermal-mesodermal boundary (Ohuchi et al., 1997). In salamanders, Fgf-8 is initially expressed in the pre-limb bud stage in the epidermis of the prospective limb bud. However, in the early limb bud stage Fgf-8 is expressed in the epidermis and mesenchyme and, as outgrowth proceeds, becomes localized to the mesenchyme and subepidermal tissue of the distal limb bud region with a stronger expression anteriorly than posteriorly (Han et al., 2001). Christensen et al. (2002) additionally investigated expression of Fgf-4 and Fgf-10 and found that only Fgf-8 and Fgf-10 were highly expressed in the developing limb but not Fgf-4, although Fgf-4 is highly expressed in the posterior portion of the AER during limb development of other tetrapods. Based on the observations that Fgf-8 and Fgf-10 are independent from the ectodermal-mesodermal interaction in salamanders and the different expression pattern of Fgf-8 in combination with the absence of an AER, they suggested that limb bud outgrowth in urodeles is not the same as in other tetrapods.

Homebox genes are essential for the regulation of the outgrowth and patterning of the limb during development and are expressed in distinct spatial and temporal patterns (e.g., Dolle et al., 1993; Nelson et al, 1996). Among those, the genes of the Hox A cluster are involved in the patterning along the proximodistal axis of the developing limb. Gardiner et al. (1995) investigated the expression patterns of Hoxa-13 and Hoxa-9 in the axolotl and found them to be essentially comparable to those of other tetrapods displaying the same temporal and spatial colinearity. Hoxa-13, involved in the patterning of distal limb structures, is expressed later and more distally than Hoxa-9, which is involved in the patterning of more proximal limb structures. Conversely, the expression of Hoxa-11 differs significantly from the pattern seen in amniotes and frogs (Wagner et al., 1999). Hoxa-11 co-expressed with Hoxa-13 is usually a marker of zeugopodial identity, with a sharp expression boundary at the distal end of the zeugopod. In urodeles, this distal expression boundary is less defined. Moreover, expression ceases while the first two digits (I and II) are formed but Hoxa-11 is re-expressed in the posteriorly forming digits (III and IV) as well as in the middle of the limb bud (Wagner et al., 1999). Therefore, digits I and II form in the absence of Hoxa-11 expression as in other tetrapods, while more postaxial digits form from Hoxa-11-expressing tissue. This expression pattern of Hoxa-11 is unique to salamanders and has not been reported for any other tetrapod taxon.

Torok et al. (1998) investigated the expression pattern of Hoxd-8, Hoxd-10, and Hoxd-11 in the axolotl and reported many similarities to the pattern reported for other tetrapods, including spatial and temporal colinearity and the general position of expression domains. Hoxd-8 is uniformly expressed in the emerging limb bud and becomes more localized distally in later stages. In the digit-forming phases of limb development, expression is weakest in the region of the forming stylopod and stronger anterodistally. Expression ceases after the formation of digits I and II.

Comparable to the pattern in other tetrapods, Hoxd-10 is not expressed in the anteroproximal region of the limb bud. When digits start to form, expression is confined to the proximal regions of the limb bud and strongest in the zeugopodial region. In the late phases of limb development, Hoxd-10 is expressed in the region of condensing zeugopodial elements and in the autopod (Torok et al., 1998).

Hoxd-11 is confined to the posterior half of the limb bud and develops into a band extending anteroposteriorly across the limb bud in later stages, but retains a stronger expression posteriorly than anteriorly. As in other tetrapods, expression is entirely absent from the most anterior cells of the developing limb. The proximal band of Hoxd-11 expression intensifies and separates from a posterodistal domain. The latter is weaker than in other tetrapods and lacks an anterior expansion. An expression-free area lies in between these two expression domains as reported for other tetrapods (e.g., Woltering and Duboule, 2010) and is the region in which the wrist will form. The cells of the posterodistal expression domain later give rise to digits II, III, and IV, while digit I forms in the absence of Hoxd-11 expression. This expression boundary between digits I and II is conserved relative to other tetrapods, although the mode of digit development of the axolotl differs significantly from that of amniotes and frogs (see above; Torok et al., 1998).

Recently, Guimond et al. (2010) investigated Bmp-2 expression in developing and regenerating limbs of Ambystoma mexicanum. They demonstrated that Bmp-2 and its target Sox-9 are independent of shh signaling in salamanders, corroborating studies on chicks that Bmp-2 is not a direct downstream target of shh involved in anterior-posterior patterning. Moreover, the expression of Bmp-2 and Sox-9 in urodeles correlates with chondrogenesis and the appearance of skeletal elements as described for other tetrapod taxa. Expression in later stages of limb development are also very similar to other tetrapods despite the absence of a paddle stage, with Bmp-2 expressed in the interdigital tissues where it was shown to be essential for giving anterior-posterior identity to digit primordia of other tetrapod taxa (Dahn and Fallon, 2000; Suzuki et al., 2008; Guimond et al., 2010).

REGENERATION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. DEVELOPMENT
  5. REGENERATION
  6. WHAT THE FOSSILS TELL US
  7. DISCUSSION
  8. Acknowledgements
  9. REFERENCES

Recognized in 1768 by Spallanzani, limb regeneration has attracted research ever since and with the advancement of molecular biology a number of the genes involved in development and regeneration of salamander limbs have been studied (see Nye et al., 2003; Gardiner and Bryant, 2007 for comprehensive reviews). A central question in the study of limb regeneration is to what extent the processes involved in the initial development of the limb and the regeneration of an extremity or parts thereof are the same and how these two processes differ. The results indicate that the patterning processes during limb development and regeneration utilize a similar set of signals after a certain point of the regeneration cascade has been reached, specifically the stage when the regenerating limb becomes independent from enervation and the blastema growth is initiated (Muneoka and Bryant, 1982; Muneoka and Sassoon, 1992; Gardiner et al., 1995; Gardiner et al., 1999). Prior to this point of convergence there are sets of events that involve “regeneration genes,” which are distinct from the processes of limb development. This includes an early phase with genes active in the wound healing process of the stump and initiation of the regeneration process, many of which are shared with other wound-healing processes and in general epidermal regeneration (Carlson et al., 1998; Gardiner et al., 1999). Among these are Msx2 in the leading edge of the epidermis, Mmp9, and members of the HoxD complex (Gardiner et al., 1995, 1999; Torok et al., 1998).

The expression of a second group of genes is dependent on the processes in the first step, although the latter do not directly induce them. However, intact wound healing and a direct contact with the epidermis are necessary for limb regeneration to proceed (Goss, 1969; Gardiner et al., 1999). To the second group belong members of the HoxA complex, namely Hoxa-9 and Hoxa-13, that are involved in the specification of the proximal-distal positions in the regenerate and the induction of the replacement of the missing parts (Gardiner et al., 1995, 1999). The formation of blastema cells through the dedifferentiation process and, therefore, a reentry into the cell cycle is a profound difference between limb development and the regeneration of limbs (Brockes, 1997). It involves essential steps, particularly in those cells that do not proliferate in their differentiated state, which are only partially understood to date (Tanaka et al., 1997). Therefore, the developmental origin of the cells in the limb bud and the blastema, respectively, before their aggregation to form the limb bud or growth zone, represent one of the fundamental differences between development and regeneration (Muneoka and Sassoon, 1992).

In limb development, the limb pattern is specified in a proximal to distal direction, with autopodial patterns being laid down last, specified by the overlapping domains of Hoxa-9 and Hoxa-13. In regeneration, expression patterns of HoxA and HoxD complexes indicate that distal parts are specified first before proximal structures are specified (Gardiner et al., 1999), because Hoxa-9 and Hoxa-13 are expressed at the same time in the amputation area independent of where the amputation took place (Gardiner et al., 1995, 1999). The different positional values of the cells that have thus been specified as distal and the cells of the stump with their original positional values stimulate intercalary growth, ultimately leading to the replacement of all missing parts of the limb (Gardiner et al., 1995, 1999). During the growth of the regenerate, the pattern of HoxA gene expression in development and regeneration converge, with Hoxa-9 expressed at the base of the blastema, and the Hoxa-13 expression more distally in the autopodial region (Gardiner et al., 1995, 1999). Beyond the transition into the nerve-dependent state of the blastema, the events following in the regeneration process (including the expression of Hoxd-11 and Shh) are basically recapitulating developmental processes (Gardiner et al., 1999). Furthermore, the limb patterning processes are also similar in developing and regenerating limbs in terms of their signaling centers: the epidermis forms an AER equivalent zone in developing limbs and the wound epidermis plays an important role in the induction of the blastema; a Shh-expressing region with polarizing influence (ZPA) is present in both the limb bud and the blastema; there is a proliferation zone in the developing limb bud and in the mesenchymal cone of the blastema (Goss, 1969; Torok et al., 1999a; Galis et al., 2003b). Finally, the limb bud and the regeneration blastema both have a high degree of self-organization capacity and can be successfully transplanted and explanted to other body regions of the same animal or other individuals (Galis et al., 2003b).

Galis et al. (2003a, b) investigated the question of why limb generation is possible in amphibians, particularly urodeles, but not in amniotes (at least under natural conditions). They proposed that although much of the development of the limb is similar in all tetrapods, the limbs of amphibians develop as a semiautonomous unit, while in amniotes, limb development primarily takes place in the phylotypic stage in which many parts of the embryo are patterned and a strong interaction between secondary fields and transient structures like the limb bud, somites, notochord, and neural tube takes place. The limb buds occur at comparable stages in all investigated developmental series of amniotes, in which the organization of the embryo is still strongly dependent on interactions between different embryonic units (Galis and Metz, 2001; Galis et al., 2003b). Under this hypothesis, a number of transient structures, which are only present in the early embryo, are necessary for limb development and these interactions cannot be repeated at a later stage, because the structures necessary for induction and interaction of limb structures have disappeared or differentiated (Galis et al., 2003b).

Although salamanders show considerable variation between the state of development of the limbs at the time of hatching (see above), limb development of most taxa is strongly delayed compared to amniote taxa. The delay of limb development relative to amniotes is even more pronounced in frogs, which have a highly modified larval stage. They lack limbs altogether as tadpoles and develop limbs only with the onset of metamorphosis (Duellman and Trueb, 1986). A delay of limb development is thought to be plesiomorphic for modern amphibians as indicated by the fossil record of Paleozoic amphibians, including branchiosaurids, which share other features of limb development with salamanders (Fröbisch et al., 2007, see below; Fröbisch, 2008). The late onset of limb development in amphibians may have been a crucial factor for the evolution of regenerative ability in salamanders and frogs, because the lag of limb development after other patterning processes in the embryo meant a freeing of the interactions with transient structures and the acquirement of a semiautonomous state of the limb (Galis et al., 2003b). Unfortunately, regeneration has not been demonstrated in the fossil taxa thus far and due to the limits of the fossil record it may be difficult to demonstrate directly. However, given the similarities in timing between salamander and branchiosaurid limb development, it seems likely that regeneration of limbs was possible in Paleozoic branchiosaurid amphibians and potentially other taxa as well.

WHAT THE FOSSILS TELL US

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. DEVELOPMENT
  5. REGENERATION
  6. WHAT THE FOSSILS TELL US
  7. DISCUSSION
  8. Acknowledgements
  9. REFERENCES

Fossils provide crucial data for studies of tetrapod limb evolution and add a deep time perspective to evolutionary hypotheses. Much can be learned on the morphological and functional evolution and diversification of the tetrapod limb from studies of the adult morphology of fossil taxa (e.g., Shubin et al., 1997; Coates et al., 2000; Carroll and Holmes, 2007; Coates and Ruta, 2007; Larsson, 2007; Shapiro et al., 2007; Frobisch and Reisz, 2009; Sigurdsen and Bolt, 2009). Unfortunately, ontogenetic series of fossil taxa that allow for investigations of at least parts of an ontogenetic trajectory are exceedingly rare and most of the time fossils only provide a snapshot of a specific ontogenetic stage.

There are some exceptions, which include fossil taxa preserved in lagerstätten, localities with extraordinary conditions for the preservation of tissues that preserve large numbers of specimens often with exquisite details of their anatomy. Of special interest in the evolution of salamander limb development is therefore the Lower Permian lake deposits of central Europe. The lagerstätten of central Europe include deposits of fossil lakes of various sizes, ranging from very large, deep lakes with multiple kilometers in diameter to rather small ponds, which have been deposited in intermountaine basins of the Variscian mountain range (Schoch, 1992; Becq-Giraudon et al., 1996; Boy and Sues, 2000). The fauna of these lakes has been exquisitely preserved, which allowed for detailed studies not only of individual lake inhabitants but also of the ecosystem dynamics in the lakes (Boy, 1998, 2003; Boy and Schindler, 2000; Schoch, 2003, 2009; Witzmann and Pfretzschner, 2003; Witzmann, 2006; Witzmann and Schoch, 2006a, b). One group of amphibians that is particularly well represented is branchiosaurids, small gill-bearing amphibians with a superficially salamander-like appearance. Hundreds of articulated branchiosaurid specimens were preserved in the fossil lake deposits, many of them with “skin shadows”, i.e., mineralized microbial mats outlining the body shape, sometimes even reflecting colouration (Willems and Wuttke, 1987; Werneburg, 2007), external gills, and stomach contents (Fig. 3A). Also preserved are a wide range of developmental stages spanning the ontogenetic trajectory from very young larvae at the onset of ossification to fully ossified adults, which provided the basis for many studies of the ossification sequence of their skulls and postcrania as well as life history pathways (Boy, 1972, 1974, 1986, 1987; Werneburg, 1991, 2001, 2002; Schoch, 1992, 2002, 2004, 2006; Schoch and Carroll, 2003; Schoch and Fröbisch, 2006; Fröbisch et al., 2007; Fröbisch and Schoch, 2009b). The branchiosaurid assemblage from one locality in the Saar-Nahe basin, the Erdesbach locality, has extraordinary taphonomic conditions and comprises over 600 specimens all deriving from a single horizon. It has provided unique insights into the ossification sequence of the limb skeleton of branchiosaurids (Fröbisch et al., 2007; Fröbisch, 2008). This revealed that branchiosaurids show a distinct preaxial dominance in the sequence of ossification of their limb skeleton, starting ossification with digit II followed by digits I, III, IV (and V in the hind limbs) (Fig. 3B). Moreover, ossification of the preaxial zeugopodial elements (radius/tibia) clearly leads in ossification before the postaxial elements (ulna/fibula). The ontogenetic series for the branchiosaurid Apateon is extraordinary in its completeness, but can only provide insights into the sequence of ossification rather than earlier events in skeletogenesis, because only bony tissues are fossilized. Nevertheless, it has been demonstrated that although ossification is not a recapitulation of the sequence of events in early skeletogenesis, the general direction and polarity of digit development is preserved in the ossification sequence, with a close correlation in particular in anamniotes (Fröbisch, 2008). The ossification sequence can therefore provide insights into the early patterning of the limbs of this extinct clade despite the fact that the sequence of chondrogenesis, being of soft tissue, cannot be preserved.

thumbnail image

Figure 3. A: Representative preservation of a Paleozoic branchiosaurid (Apateon) from the Saar-Nahe basin in Western Germany. B: Graph depicting the sequence of ossification in the limbs of selected tetrapod taxa with ontogenetic rank plotted versus ossification event. Note the similar or identical course of the curve for the branchiosaurid Apateon and extant salamanders, contrasted by the divergent course pattern in amniotes and frogs, reflecting the reversed polarity in digit ossification of Apateon and salamanders relative to amniotes and frogs (modified from Fröbisch et al., 2007). C: Pes of the amphibamid Gerobatrachus hottoni, showing an ossified basale commune (modified from Anderson et al., 2008).

Download figure to PowerPoint

Branchiosaurids are the only tetrapod group outside of urodeles for which preaxial polarity of digit formation has been demonstrated. It remains unknown if branchiosaurids also possessed a basale commune, because their carpals and tarsal remained unossified and hence were not preserved. However, a basale commune has been identified in a close relative of branchiosaurids, the amphibamid Gerobatrachus hottoni, which unites ancient features of Paleozoic tetrapods with features seen in modern frogs and salamanders (Anderson et al., 2008; see Marjanoviç and Laurin, 2008, for a different interpretation) (Fig. 3C).

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. DEVELOPMENT
  5. REGENERATION
  6. WHAT THE FOSSILS TELL US
  7. DISCUSSION
  8. Acknowledgements
  9. REFERENCES

Until recently, the distinctive features of salamander limb development including the preaxial polarity in digit formation and the presence of a basale commune were considered autapomorphic for Urodela (Vorobyeva and Hinchliffe, 1996; Wake and Shubin, 1998; Wagner et al., 1999; Shubin and Wake, 2003; Stopper and Wagner, 2007). In fact, in the first hypothesis put forward to explain the aberrant urodele pattern, Holmgren (1933) and later Jarvik (1942, 1965) considered the differences between patterning of the limb skeleton of salamanders and all other tetrapods so profound that they proposed a dual origin of tetrapods. Therein, Holmgren (1933) proposed a relationship between Dipnoi and urodeles (Fig. 4), while Jarvik (1942, 1965) supported a urodele-porolepiform relationship. However, nowadays the monophyly of Tetrapoda is uncontested and solidly rests on a vast amount of molecular and morphological data. Hence other evolutionary scenarios have been put forward to explain the divergence between the two patterns of tetrapod limb development. A hypothesis proposed by Wagner et al. (1999) was based on their observations that Hoxa-11 is expressed during development of the postaxial digits of salamanders (see above). They considered their molecular data within a classic phylogenetic framework in which salamander taxa with reduced digits are placed basal within the salamander phylogeny. This let them to propose that salamanders had undergone an evolutionary phase of digit reduction in which the stem lineage of salamanders reduced their digits to two (Fig. 5). Subsequently, in the evolutionary history of urodeles, the digits on the posterior side of the salamander autopodium evolved as de novo structures. Hence, salamander digits I and II would be homologues of digits III and IV of other tetrapods, but the posterior digits of the salamander autopodium are not homologous to the digits of other tetrapods (Fig. 5). In the framework of their molecular data alone, this evolutionary scenario seemed logical and plausible. However, recent analyses of higher level salamander phylogenies have yielded controversial results. While a comprehensive and well-supported analysis by Zhang and Wake (2009) based on mitochondrial DNA supports the position of sirenids as sister taxon to all other salamanders as did earlier studies (Larson and Dimmick, 1993), other analyses have yielded cryptobranchids and hynobiids as the basalmost salamanders (Wiens et al., 2005, Roelants et al, 2007; Vietes et al., 2009) and a more derived position of taxa with reduced limbs, i.e., sirenids, amphiumids, and proteiids, indicating that limb reduction may not be a plesiomorphic trait in salamanders. Furthermore, the fossil record does not support the proposed phase of digit reduction in salamander evolution since neither any of the potential Permo-Carboniferous relatives of salamanders, nor the earliest crown-group salamanders recovered from Jurassic and Cretaceous deposits in China show a tendency towards digit reduction (Gao and Shubin, 2001, 2003; Carroll and Holmes, 2007).

thumbnail image

Figure 4. Phylogenetic relationships within Sarcopterygia as envisioned in the hypothesis for a diphyletic origin of tetrapods.

Download figure to PowerPoint

thumbnail image

Figure 5. Hypothesis of a neomorphic origin of postaxial digits in salamanders as proposed by Wagner et al. (1999). Salamanders (1) undergo an early evolutionary phase of digit reduction, where preaxial digits I and II are reduced and only postaxial digits III and IV remain (2). Subsequently, new postaxial digits re-evolve (3). Therefore, preaxial digits in the salamander autopod are homologous of digits III and IV in other tetrapods, while the neomorphic postaxial digits are not homolgous to other tetrapod digits.

Download figure to PowerPoint

The recent discoveries of preaxial dominance in limb development in the Lower Permian branchiosaurids (Fröbisch et al., 2007; Fröbisch, 2008) and a basale commune in the Middle Permian amphibamid Gerobatrachus (Anderson et al., 2008) add further data points for the evolution of this developmental pathway deep in the evolutionary history of amphibians (Fig. 6). This is of particular interest when considered in the framework of the controversial origin(s) and relationships of modern amphibians for which three scenarios are being discussed in the literature: the polyphyly hypothesis proposes a sister taxon relationship between salamanders and branchiosaurids, and frogs and amphibamids, respectively, while caecilians are considered most closely related to microsaurs within lepospondyls (e.g., Carroll, 2007; Anderson et al., 2008; Anderson, 2008;). In the most widely accepted temnospondyl hypothesis, modern amphibians form a monophyletic clade (“Lissamphibia”) nested within temnospondyls and most closely related to amphibamids or branchiosaurids (e.g., Bolt, 1991; Milner, 1993; Ruta et al., 2003; Schoch and Milner, 2004; Ruta and Coates, 2007; Sigurdsen and Bolt, 2009). In the lepospondyl hypothesis, frogs, salamanders, and caecilians also form a monophyletic Lissamphibia with lysorophian lepospondyls as their closest relatives (Laurin and Reisz, 1997; Vallin and Laurin, 2004; Marjanovic and Laurin, 2007, 2008). Branchiosaurids and amphibamids have classically played a central role in this controversy with both of the groups favored as close relatives to some or all of the modern amphibians. However, the presence of a basale commune alone does not automatically imply preaxial dominance in limb development and an ontogenetic series of Gerobatrachus is currently not available that could unequivocally demonstrate preaxial dominance in limb development. Likewise, preaxial dominance in limb development was observed in branchiosaurids, but the carpals and tarsals of branchiosaurids remained unossified and, therefore, unfossilized and hence direct evidence for a basale commune in branchiosaurids is lacking. Nonetheless, a recent comprehensive phylogenetic analysis of branchiosaurids and amphibamids showed that branchiosaurids are actually a monophyletic subclade within amphibamids (Fröbisch and Schoch, 2009a), demonstrating that these apomorphic features of limb development, i.e., preaxial dominance and a basale commune, are found within a single evolutionary lineage and it seems likely that both features have evolved in concert (Fig. 6).

thumbnail image

Figure 6. Hypothesis of preaxial polarity in digit development as a larval adaptation under consideration of the fossil data. Preaxial polarity in digit development is known from modern salamanders as well as Paleozoic branchiosaurids. A basale commune is known from modern salamanders and the Paleozoic amphibamid Gerobatrachus. Assuming monophyly of modern amphibians, preaxial polarity in limb development (A) may have evolved convergently in salamanders and amphibamids (including branchiosaurids), or is shared by the clade of amphibamids + Lissamphibia and is reversed to postaxial polarity in digit development (B) in frogs. Phylogenetic relationships and stratigraphic ranges combined from several sources (Holmes, 2000; Milner, 2000; Wiens et al., 2005; Fröbisch and Schoch, 2009a).

Download figure to PowerPoint

The currently most widely accepted hypothesis to explain the evolution of preaxial dominance in limb development is that it arose as a larval adaptation facilitating interaction of the free swimming larva with the environment while limbs are developing (Schmalhausen, 1910; Vorobyeva and Hinchliffe, 1996; Shubin and Wake, 2003). Given the data at hand, it may be tempting to hypothesize that preaxial dominance in limb development has evolved only once in the lineage leading to modern salamanders and that this lends support for a close relationship with amphibamids (including branchiosaurids), which share many aspects of their larval and adult ecology with modern salamanders (Boy and Sues, 2000; Werneburg, 2002; Schoch and Fröbisch, 2006; Fröbisch and Schoch, 2009b). However, these Paleozoic taxa are separated by roughly 100 million years from the first representatives of crown-group salamanders in the fossil record, a critical period from which essentially no data for salamander evolution is available. Moreover, when the character of preaxial dominance in limb development is considered in a phylogenetic framework, it becomes clear that other evolutionary scenarios are only slightly less parsimonious, including preaxial dominance as the plesiomorphic state for all tetrapods, retained only in salamanders among living tetrapods, or a convergent evolution of preaxial dominance in amphibamids (including branchiosaurids) and urodeles (Fröbisch et al., 2007).

The evolution and phylogeny of crown group salamanders is plagued by homoplasy. In fact, a large a number of highly derived anatomical characters, including body elongation, tail autonomy, and life history pathways, have been demonstrated or are debated to have evolved multiple times (Wake, 1991; Chippindale et al., 2004; Mueller et al., 2004; Bonett et al., 2005; Bruce, 2005; Chippindale and Wiens, 2005). Phylogenetic analyses based on molecular data and combined datasets have played a crucial role in identifying these homoplasies, where morphology alone is often misleading. In particular, the evolution of life history strategies, i.e., larval development versus direct development (Chippindale et al., 2004; Mueller et al., 2004; Bonett et al., 2005; Bruce, 2005; Chippindale and Wiens, 2005), are also of interest for discussions on the evolution of salamander limb development, because these life history strategies are known to be associated with distinct patterns and timing of digit development (Shubin and Wake, 2003; Franssen et al., 2005). If direct development evolved multiple times within plethodontid salamanders and/or if larval development re-evolved in certain clades within plethodontids, the question arises whether associated patterns and timing of digit development and the role of apoptosis in autopod development also (re-)evolved multiple times in concert with these life history strategies. Currently, Desmognathus aenus is the only direct developing salamander for which a detailed study of limb development has been conducted (Franssen et al., 2005) and a much more diverse sampling of direct developers and larval developers among urodeles will be necessary to answer these questions.

A current phylogenetic framework and the fossil record helped to elucidate that molecular developmental data taken by itself may be misleading and hopefully future fossil finds will continue to fill the gaps in our knowledge on salamander evolution (Fig. 7). In turn, it is now essential to gain a better understanding of the molecular control of salamander limb development to understand the processes involved in the patterning of urodele limbs and to get a better idea of the complexity of this developmental pathway and thereby its phylogenetic history and the likelihood of its convergent evolution (Fig. 7).

thumbnail image

Figure 7. Summary of the changing hypotheses for the evolution of preaxial polarity in tetrapod limb development, showing how the consideration of different lines of evidence, i.e., morphological, developmental, phylogenetic, and fossil data, has led to vastly different scenarios and changing phylogenetic scales for the evolution of this developmental pathway.

Download figure to PowerPoint

Integrative approaches have previously successfully been taken to elucidate the evolutionary history of complex morphologies, e.g., the initial evolution of the tetrapod autopodium (e.g., Shubin and Alberch, 1986; Shubin et al., 1997; Shubin, 2002; Larsson, 2007; Shubin et al., 2009) or the evolution of the avian wing from a theropod autopod (e.g., Wagner and Gauthier, 1999; Larsson and Wagner, 2002; Galis et al., 2003a; Wagner, 2005). The case of salamander limb development highlights that an integration of molecular approaches, data from the fossil record, and a phylogenetic framework is essential to gain a comprehensive picture for the evolution of patterns and processes in tetrapod limb development and evolution.

Acknowledgements

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. DEVELOPMENT
  5. REGENERATION
  6. WHAT THE FOSSILS TELL US
  7. DISCUSSION
  8. Acknowledgements
  9. REFERENCES

NSF Grant 0544565 was awarded to N.H.S. and a Research Fellowship, from the Deutsche Forschungsgemeinschaft (DFG) was awarded to N.B.F.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. DEVELOPMENT
  5. REGENERATION
  6. WHAT THE FOSSILS TELL US
  7. DISCUSSION
  8. Acknowledgements
  9. REFERENCES
  • Alberch P, Gale EA. 1985. A developmental analysis of an evolutionary trend: digital reduction in amphibians. Evolution 39: 823.
  • Anderson JS. 2008. Focal review: the origin(s) of modern amphibians. Evol Biol 35: 231247.
  • Anderson JS, Reisz RR, Fröbisch NB, Scott D, Sumida SS. 2008. A stem batrachian from the Early Permian of Texas and the origin of frogs and salamanders. Nature 453: 515518.
  • Becq-Giraudon J-F, Montenant C, Van den Driessche J. 1996. Hercynian high-altitude phenomenan the French Massif Centrale: tectonic implications. Paleogeogr Paleoclimatol Paleoecol 122.
  • Bininda-Emonds ORP, Jeffery JE, Sanchez-Villagra MR, Hanken J, Colbert M, Pieau C, Selwood L, ten Cate C, Raynaud A, Osabutey CK, Richardson MK. 2007. Forelimb-hind limb developmental timing across tetrapods. BMC Evol Biol.
  • Blanco MJ, Alberch P. 1992. Caenogenesis, developmental variability, and evolution in the carpus and tarsus of the marbled newt Triturus marmoratus. Evolution 46: 677687.
  • Bolt JR. 1991. Lissamphibian origins. In: SchultzeH-P, TruebL, editors. Origins of the higher groups of tetrapods: controversy and consensus. Ithaca, NY: Comstock Pub. Associates. p 194221.
  • Bonett RM, Mueller RL, Wake DB. 2005. Why should reacquisition of larval stages by desmognathine salamander suprise us? Herpetol Rev 36: 112113.
  • Boy JA. 1972. Die Branchiosaurier (Amphibia) des saarpfaelzischen Rotliegenden (Perm, SW-Deutschland). [Branchiosaurids (Amphibia) of the Palatine Rotliegend (Permian, SW Germany).] Abhandlung hessischen Landesamt Bodenforschung 65: 6137.
  • Boy JA. 1974. Die Larven der rhachitomen Amphibien (Amphibia: Temnospondyli; Karbon-Trias). [The larvae of rhachitomous amphibians (Amphibia: Temnospondyli; Carboniferous-Triassic).] Paläontol Z 48: 236268.
  • Boy JA. 1986. Studien über die Branchiosauridae (Amphibia: Temnospondyli) 1. Neue und wenig bekannte Arten aus dem mitteleuropäischen Rotliegenden (oberstes Karbon bis unteres Perm). [Studies on Branchiosauridae (Amphibia:Temnospondyli) 1. New and poorly known species from the middle European Rotliegend (Upper Carboniferous to Lower Permian).] Paläontol Z 60: 131166.
  • Boy JA. 1987. Studien über die Branchiosauridae (Amphibia: Temnospondyli); Ober-Karbon - Unter-Perm. Systematische Übersicht. [Studies on Branchiosauridae (Amphibia:Temnospondyli); Upper Carboniferous - Lower Permian. Systematic overview.] N Jahrbuch Geol Paläontol Abhandlungen 174: 75104.
  • Boy JA. 1998. Möglichkeiten und Grenzen einer Ökosystem-Rekonstruktion am Beispiel des lakustrinen Paläo-Ökosystems. 1. Theoretische und methodische Grundlagen. [Possibilities and limits of ecosystem-reconstruction using the example of of lacustrine paleo-ecosystems. 1. Theoretical and methodological basics]. Paläontol Z 72: 207240.
  • Boy JA. 2003. Paläoökologische Rekonstruktion von Wirbeltieren: Möglichkeiten und Grenzen. [Paleoecological reconstruction of vertebrates: possibilities and limits.] Paläontol Z 77: 123152.
  • Boy JA, Schindler T. 2000. Ökostratigraphische Bioevents im Grenzbereich Stephanium/Autunium (höchstes Karbon) des Saar-Nahe-Beckens (SW-Deutschand) und benachbarter Gebiete. [Ecostratigraphic bioevents at the Stephanian-Autunian boundary (Upper Carboniferous) of the Saar-Nahe basin (SW Germany) and neighboring areas.] N Jahrbuch Geol Paläontol Abhandlungen 216: 89152.
  • Boy JA, Sues H-D. 2000. Branchiosaurs: Larvae, metamorphosis and heterochrony in temnospondyls and seymouriamorphs. In: HeatwoleH, CarrollRL, editors. Amphibian biology. Chipping Norton: Surrey Beatty & Sons PTY Limited. p 11501197.
  • Brockes JP. 1997. Amphibian limb regerneration: rebuilding a complex structure. Science 276: 8187.
  • Bruce RC. 2005. Did Desmognathus salamanders reinvent the larval stage? Herpetol Rev 36: 107112.
  • Bryant SV, Muneoka K. 1986. Views of limb development and regeneration. Trends Genet 2: 153159.
  • Carlson MRJ, Bryant SV, Gardiner DM. 1998. Expression of Msx-2 during development, regeneration, and wound healing in axolotl limbs. J Exp Zool 282: 715723.
  • Carroll RL. 2007. The Palaeozoic ancestry of salamanders, frogs, and caecilians. Zool J Linn Soc 150( Suppl 1): 1140.
  • Carroll RL, Holmes RB. 2007. Evolution of the appendicular skeleton of amphibians. In: HallBK, editor. Fins into limbs. Chicago: University of Chicago Press. p 185224
  • Chippindale PT, Wiens JJ. 2005. Re-evolution of the larval stage in the plethodontid salamander genus Desmogathus. Herpetol Rev 26: 113117.
  • Chippindale PT, Bonett RM, Baldwin AS, Wiens JJ. 2004. Phylogenetic evidence for a major reversal of life-history evolution in plethodontid salamanders. Evolution 58: 28092822.
  • Christensen RN, Weinstein M, Tassava RA. 2002. Expression of fibroblast growth factors 4, 8, and 10 in limbs, flanks, and blastemas of Ambystoma. Dev Dyn 223: 193203.
  • Coates MI, Ruta M. 2007. Skeletal changes in the transition from fins to limbs. In: HallBK, editor. Fins into limbs. Chicago: University of Chicago Press. p 1537.
  • Coates MI, Ruta M, Milner AR. 2000. Early tetrapod evolution. Trends Ecol Evol 15: 327328.
  • Dahn RD, Fallon JF. 2000. Interdigital regulation of digit identity and homeotic transformation by modulated BMP signaling. Science 289: 438441.
  • Dolle P, Dierich A, LeMeur M, Schimmang T, Schuhbaur B, Chambon P, Duboule D. 1993. Disruption of the Hoxd13 gene induces localized heterchrony leading to mice with neotenic limbs. Cell 75:431441.
  • Duellman WE, Trueb L. 1986. Biology of amphibians. New York: McGraw-Hill Book Co.
  • Erdmann K. 1933. Zur Entwicklung des knöchernen Skelets von Triton und Rana unter besonderer Berücksichtigung der Zeitfolge der Ossifikationen. [On the development of the bony skeleton of Triton and Rana under special consideration of the order of ossification.] Z gesamte Anat I. Abt. 101: 566649.
  • Franssen RA, Marks S, Wake D, Shubin N. 2005. Limb chondrogenesis of the seepage salamander, Desmognathus aeneus (Amphibia: Plethodontidae). J Morphol 265: 87101.
  • Frobisch J, Reisz RR. 2009. The Late Permian herbivore Suminia and the early evolution of arboreality in terrestrial vertebrate ecosystems. Proc R Soc B-Biol Sci 276: 36113618.
  • Fröbisch NB. 2008. Ossification patterns in the tetrapod limb: conservation and divergence from morphogenetic events. Biol Rev 83: 571600.
  • Fröbisch NB, Schoch RR. 2009a. Testing the impact of miniaturization on phylogeny: paleozoic dissorophoid amphibians. System Biol 58: 312327.
  • Fröbisch NB, Schoch RR. 2009b. The largest specimen of Apateon and the life history pathway of neoteny in the Paleozoic temnospondyl family Branchiosauridae. Fossil Rec 12: 8390.
  • Fröbisch NB, Carroll RL, Schoch RR. 2007. Limb ossification in the Paleozoic branchiosaurid Apateon (Temnospondyli) and the early evolution of preaxial dominance in tetrapod limb development. Evol Dev 9: 6975.
  • Galis F, Metz JAJ. 2001. Testing the vulnerability of the phylotypic stage: on modularity and evolutionary conservation. J Exp Zool 41: 195204.
  • Galis F, Kundrat M, Sinervo B. 2003a. An old controversy solved: bird embryos have five fingers. Trends Ecol Evol 18: 79.
  • Galis F, Wagner PG, Jockusch EL. 2003b. Why is limb regeneration possible in amphibians but not in reptiles, birds, and mammals? Evol Dev 5: 208220.
  • Gao K-Q, Shubin NH. 2001. Late Jurassic salamanders from northern China. Nature 410: 574577.
  • Gao K-Q, Shubin NH. 2003. Earliest crown-group salamanders. Nature 422: 424428.
  • Gardiner DM, Blumberg B, Komine Y, Bryant SV. 1995. Regulation of HoxA expression in developing and regenrating axolotl limbs. Development 121: 17311741.
  • Gardiner DM, Carlson MRJ, Roy S. 1999. Towards a functional analysis of limb regeneration. Cell Dev Biol 10: 385393.
  • Gardiner DM, Bryant SV. 2007. Tetrapod limb regeneration. In: HallBK, editor. Fins into limbs. Chicago: University of Chicago Press. p 163182.
  • Goss RJ. 1969. Principles of regeneration. New York: Academic Press. 278p.
  • Gould SJ. 1977. Ontogeny and phylogeny. Cambridge: The Belknap Press of Harvard University. 501 p.
  • Guimond J-C, Levesque M, Michaud P-L, Berdugo J, Finnson K, Philip A, Roy S. 2010. BMP-2 functions independently of SHH signaling and triggers cell condensation and apoptosis in regenerating axolotl limbs. BMC Dev Biol 10: 15.
  • Hall BK. 2007. Fins into limbs. Chicago: The University of Chicago Press. 433 p.
  • Han M-J, An J-Y, Kim W-S. 2001. Expression pattern of Fgf-8 during development and limb regeneration of the axolotl. Dev Dyn 220: 4048.
  • Holmes R. 2000. Paleozoic temnospondyls. In: HeatwoleH, CarrollRL, editors. Amphibian biology. Vol. 4. Chipping Norton: Surrey Beatty & Sons. p 10811120.
  • Holmgren N. 1933. On the origin of the tetrapod limb. Acta Zool 14: 185295.
  • Holmgren N. 1939. Contribution to the question of the origin of the tetrapod limb. Acta Zool : 89124.
  • Holmgren N. 1942. On the tetrapod limb: again. Acta Zool 30: 485508.
  • Howes LD, Swinnerton HH. 1900. On the development of the skeleton of the tuatara, Sphenodon punctatus. Trans Zool Soc Lond 16: 186.
  • Imokawa Y, Yoshizato K. 1997. Expression of Sonic hedgehog gene in regenerating newt limb blastemas recapitulates that in developing limb buds. Proc Natl Acad Sci 94: 91599164.
  • Jarvik E. 1942. On the structure of the snout of crossopterygians and lower gnathostomes in general. Zool Bidrag 21:235675.
  • Jarvik E. 1965. On the origin of girdles and paired fins. Israel J Zool 14: 141172.
  • Keller R. 1946. Morphogenetische Untersuchungen am Skelett von Sirenodon mexicanus Shaw mit besonderer Berücksichtigung des Ossifikationsmodus beim neotenen Axolotl. [Morphogenetic studies on the skeleton of Sirenodon mexicanus Shaw with special consideration of the mode of ossification in the neotenic axolotl.] Rev Suisse Zool 53: 329426.
  • Larson A, Dimmick WW. 1993. Phylogenetic relationships of the salamander families: an analysis of congruence among morphological and molecular characters. Herpetol Monogr 7: 7793.
  • Larsson HCE. 2007. MODEs of developmental evolution: an example with the origin and definition of the autopodium. In: AndersonJS, SuesH-D, editors. Major transitions in vertebrate evolution. Bloomington: Indiana University Press. p 150181.
  • Larsson HCE, Wagner GP. 2002. Pentadactyl ground state of the avian wing. J Exp Zool 294: 146151.
  • Laurin M, Reisz RR. 1997. A new perspective on tetrapod phylogeny. In: SumidaSS, MartinKL, editors. Amniote origins. New York: Academic Press. p 959.
  • Marjanovic D, Laurin M. 2007. Fossils, molecules, divergence times, and the origin of lissamphibians. System Biol 56: 369388.
  • Marjanovic D, Laurin M. 2008. A reevaluation of the evidence supporting an unorthodox hypothesis on the origin of extant amphibians. Contrib Zool 77: 149199.
  • Milner AR. 1993. The Paleozoic relatives of lissamphibians. Herpetol Monogr 7: 827.
  • Milner AR. 2000. Mesozoic and tertiary caudata and albenerpetontidae. In: HeatwoleH, CarrollRL, editors. Amphibian biology. Chipping Norton: Surrey Beatty & Sons. p 14121444.
  • Mueller RL, Macey JR, Jaekel M, Wake DB, Boore JL. 2004. Morphological homoplasy, life history evolution, and historical biogeography of plethodontid salamanders inferred from complete mitochondrial genomes. Proc Natl Acad Sci USA 101: 1382013825.
  • Muneoka K, Bryant SV. 1982. Evidence that patterning mechanisms in developing and regenerating limbs are the same. Nature 298: 369371.
  • Muneoka K, Sassoon D. 1992. Molecular aspects of regeneration in developing vertebrate limbs. Dev Biol 152: 3749.
  • Nelson CE, Morgan BA, Burke AC, Laufer E, DiMambro E, Murtaugh LC, Gonzales E, Tessarollo L, Parada LF, Tabin C. 1996. Analysis of Hox gene expression in the chick limb bud. Development 122:14491466.
  • Nye HL, Cameron JA, Chernoff EA, Stocum DL. 2003. Regeneration of the urodele limb: a review. Dev Dyn 226: 280294.
  • Ohuchi H, Nakagawa T, Yamamoto A, Araga A, Ohata T, Ishimaru Y, Yoshioka H, Kuwana T, Nohno T, Yamasaki M, Itoh N, Noji S. 1997. The mesenchymal factor, FGF10, initiates and maintains the outgrowth of the chick limb bud through interaction with FGF8, an apical ectodermal factor. Development 124: 22352244.
  • Oster GF, Shubin NH, Murray JD, Alberch P. 1988. Evolution and morphogenetic rules: the shape of the vertebrate limb in ontogeny and phylogeny. Evolution 42: 862884.
  • Pée Pv. 1904. Les membres chez Amphiuma. [The limbs of Amphiuma.] Anat Anzeig 24: 476482.
  • Roelants K, Gower DJ, Wilkinson M, Loader SP, Biju SD, Guillaume K, Moriau L, Bossuyt F. 2007. Global patterns of diversification in the history of modern amphibians. Proc Natl Acad Sci USA 104: 887892.
  • Ruta M, Coates MI. 2007. Dates, nodes and character conflict: addressing the lissamphibian origin problem. J System Palaeontol 5: 69122.
  • Ruta M, Coates MI, Quicke DLJ. 2003. Early tetrapod relationships revisited. Biol Rev 78: 251345.
  • Schmalhausen I. 1917. On the extremities of Ranodon sibiricus Kessl. Rev Zool Russe 2: 129135.
  • Schmalhausen JJ. 1907. Die Entwicklung des Sklettes der vorderen Extremitaet der anuren Amphibien. [Development of the forelimb skeleton of anurans.] Anat Anzeig 31: 177187.
  • Schmalhausen JJ. 1910. Die Entwicklung des Extremitätenskelettes von Salamandrella kayserlingii. [Development of the limb skeleton of Salamandrella keyserlingii.] Anat Anzeig 37: 431446.
  • Schoch RR. 1992. Comparative ontogeny of early Permian branchiosaurid amphibians from southwestern Germany. Palaeontographica, Abt. A 222: 4383.
  • Schoch RR. 2002. The early formation of the skull in extant and Paleozoic amphibians. Paleobiology 28: 278296.
  • Schoch RR. 2003. Early larval ontogeny of the Permo-Carboniferous Temnospondyl Sclerocephalus. Paleontology 46: 118.
  • Schoch RR. 2004. Skeleton formation in the Branchiosauridae: A case study in comparing ontogenetic trajectories. J Vertebr Paleontol 24: 309319.
  • Schoch RR. 2006. Skull ontogeny: developmental patterns of fishes conserved across major tetrapod clades. Evol Dev 8: 524536.
  • Schoch RR. 2009. Life-cycle evolution as response to diverse lake habitats in Paleozoic amphibians. Evolution 63: 27382749.
  • Schoch RR, Carroll RL. 2003. Ontogenetic evidence for the Paleozoic ancestry of salamanders. Evol Dev 5: 314324.
  • Schoch RR, Fröbisch NB. 2006. Metamorphosis and neoteny: alternative pathways in an extinct amphibian clade. Evolution 60: 14671475.
  • Schoch RR, Milner AR. 2004. Structure and implications of theories on the origins of lissamphibians. In: ArratiaG, WilsonMVH, editors. Recent advances in the origin and early radiation of vertebrates. Munich: Verlag Dr. Friedrich Pfeil. p 345377.
  • Shapiro MD, Shubin NH, Downs JP. 2007. Limb diversity and digit reduction in reptilian evolution. In: HallBK, editor. Fins into limbs. Chicago: University of Chicago Press.
  • Shubin NH. 2002. Origin of evolutionary novelty: examples from limbs. J Morphol 252: 1528.
  • Shubin NH, Alberch P. 1986. A morphogenetic approach to the origin and basic organisation of the tetrapod limb. Evol Biol 20: 319387.
  • Shubin NH, Wake DB. 2003. Morphological variation, development, and evolution of the limb skeleton of salamanders. In: HeatwoleH, DaviesM, editors. Amphibian biology. Chipping Norton: Surrey Beatty & Sons PTY Limited. p 17821808.
  • Shubin NH, Tabin C, Carroll SB. 1997. Fossils, genes and the evolution of animal limbs. Nature 388: 639648.
  • Shubin NH, Tabin C, Carroll S. 2009. Deep homology and the origins of evolutionary novelty. Nature 457: 818823.
  • Sigurdsen T, Bolt JR. 2009. The lissamphibian humerus and elbow joint, and the origins of modern amphibians. J Morphol 270: 14431453.
  • Steiner H. 1921. Hand und Fuss der Amphibien, ein Beitrag zur Extremitaetenfrage. [Hand and foot of amphibians, a contribution to the limb question.] Anatom Anzeig 53: 513542.
  • Steiner H. 1934. Ueber die embryonale Hand- und Fuss-Skelett-Analge bei den Crocodiliern, sowie über ihre Beziehungen zur Vogel-Flügelanlage und zur ursprünglichen Tetrapoden-Extramität. [On the embryological anlage of the hand and foot skeleton in crocodiles and their relation to the analage of the bird wing and the ancient tetrapod limb.] Rev Suisse Zool 41: 383396.
  • Stopper GF, Wagner PG. 2007. Inhibition of Sonic hedgehog signalling leads to posterior digit loss in Ambystoma mexicanum: parallels to natural digit reduction in urodeles. Dev Dyn 236: 321331.
  • Strasser H. 1879. Zur Entwicklung der Extremitätenknorpel bei Salamandern und Tritonen. Morphol Jahrbuch 5: 240315.
  • Suzuki T, Hasso SM, Fallon JF. 2008. Unique SMAD1/5/8 activity at the phalanx-forming region determines digit identity. PNAS 105: 41854190.
  • Tanaka EM, Gann AAF, Gates PB, Brockes JP. 1997. Newt myotubules re-enter the cell cycle by phosphorylation of the retinoblastoma protein. J Cell Biol 136: 155165.
  • Tank PW, Carlson BM, Connelly TG. 1977. A scanning electron microscopic comparison of the development of embryonic and regenerating limbs in the axolotl. J Exp Zool 201: 417429.
  • Torok M, Gardiner DM, Shubin NH, Bryant SV. 1998. Expression of HoxD genes in developing and regenerating axolotl limbs. Dev Biol 200: 225233.
  • Torok M, Gardiner DM, Izpisua-Belmonte J-C, Bryant SV. 1999a. Sonic Hedgehog (Shh) expression in developing and regnerating axolotl limbs. J Exp Zool 284: 197206.
  • Torok MA, Gardiner DM, Izpisúa-Belmonte JC, Bryant SV. 1999b. Sonic Hedgehog (shh) expression in developing and regenerating axolotl limbs. J Exp Zool 284: 197206.
  • Vallin G, Laurin M. 2004. Cranial morphology and affinities of Microbrachis, and a reappraisal of the phylogeny and lifestyle of the first amphibians. J Vertebr Paleontol 24: 5672.
  • Vieites DR, Zhang P, Wake DB. 2009. Salamanders (Caudata). In: Hedges, SB, KumarS, editors. The timetree of life. New York: Oxford University Press. p 365368.
  • Vorobyeva EI, Hinchliffe RJ. 1996. Developmental pattern and morphology of Salamandrella keyserlingii limbs (Amphibia, Hynobiidae) including some evolutionary aspects. Russ J Herpetol 3: 6881.
  • Vorobyeva EI, Antipenkova TP, Kolobayeva OV, Hinchliffe RJ. 2000. Some peculiarities of development in two populations of Salamandrella keyserlingii (Hynobiidae, Caudata). Russ J Herpetol 7: 115122.
  • Wagner G. 2005. The developmental evolution of avian digit homology: an update. Theory Biosci 124: 165183.
  • Wagner GP, Gauthier JA. 1999. 1,2,3 = 2,3,4: A solution to the problem of the homology of the digits in the avian hand. Proc Natl Acad Sci USA 96: 51115116.
  • Wagner GP, Khan PA, Blanco MJ, Misof B, Liversgae RA. 1999. Evolution of Hoxa-11 expression in amphibians: Is the urodele autopodium an innovation? Am Zool 39: 686694.
  • Wake DB. 1991. Homoplasy: the result of natural selection, or evidence of design limitations? Am Natural 138: 543567.
  • Wake DB, Shubin NH. 1998. Limb development in the Pacific giant salamander, Dicamtodon (Amphibia, Caudata, Dicamptodontidae). Can J Zool 76: 20582066.
  • Werneburg R. 1991. Die Branchiosaurier aus dem Unterrotliegend des Döhlener Beckens bei Dresden. [Branchiosaurids from the Lower Rotliegend of the Döhlen basin near Dresden.] Veröffentlichung Naturhistorisch Museum Schleusingen 6: 7599.
  • Werneburg R. 2001. Apateon dracyiensis - eine frühe Pionierform der Branchiosaurier aus dem europäischen Rotliegend, Teil1: Morphologie. [Apateon dracyiensis - an early pioneer form of branchiosaurids from the European Rotliegend, part 1: morphology.] Veröffentlichung Naturhistorisch Museum Schleusingen 16: 1736.
  • Werneburg R. 2002. Apateon dracyiensis - eine frühe Pionierform aus dem europäischen Rotliegend, Teil 2: Paläoökologie. [Apateon dracyiensis - an early pioneer form of branchiosaurids from the European Rotliegend, part2: paleoecology.] Veröffentlichungen des Naturhistorischen Museum Schleusingen 17: 1732.
  • Werneburg R. 2007. Timeless design: colored pattern of skin in early Permian branchiosaurids (Temnospondyli: Dissorophoidea). J Vertebr Paleontol 27: 10471050.
  • Wiedersheim R. 1879. Zur Anatomie des Amblystoma weismanni. [On the natomy of Amblystoma weismanni.] Z Wissenschatliche Zool 32: 216236.
  • Wiens JJ, Bonett RM, Chippindale PT. 2005. Ontogeny discombobulates phylogeny. System Biol 54: 91110.
  • Willems H, Wuttke M. 1987. Lithogenese lakustriner Dolomite und mikrobiell induzierte Weichteilerhaltung bei Tetrapoden des Unter-Rotliegenden (Perm, Saar-Nahe Becken, SW-Deutschland). [Lithogenesis of lacustrine dolomites and microbially induced soft tissue preservation in tetrapods of the Lower Rotliegend (Permian, Saar-Nahe basin, SW-Germany).] N Jahrbuch Geol Paläontol Abhandlungen 174: 213238.
  • Witzmann F. 2006. Developmental patterns and ossification sequence in the Permo-Carboniferous temnospondyl Archegosaurus decheni (Saar-Nahe Basin, Germany). J Vertebr Paleontol 26: 717.
  • Witzmann F, Pfretzschner H-U. 2003. Larval ontogeny of Micromelerpeton credneri (Temnospondyli, Dissorophoidea). J Vertebr Paleontol 23: 750768.
  • Witzmann F, Schoch RR. 2006a. Skeletal development of the temnospondyl Acanthostomatops vorax from the Lower Permian Döhlen Basin of Saxony. Transact R Soc Edinburgh: Earth Sci 96: 365385.
  • Witzmann F, Schoch RR. 2006b. The postcranium of Archegosaurus decheni, and a phylogenetic analysis of temnospondyl postcrania. Palaeontology 49: 12111235.
  • Woltering JM, Duboule D. 2010. The origin of digits: expression patterns versus regulatory mechanisms. Dev Cell 18: 526532.
  • Zeller R, Lopez-Rios J, Zuniga A. 2009. Vertebrate limb bud development: moving towards integrative analysis of organogenesis. Nate Rev Genet 10: 845858.
  • Zhang, P, Wake DB. 2009. Higher-level salamander relationships and divergence dates inferred from complete mitochondrial genomes. Mol Phylogenet Evol 53: 492508.
  • Zwick W. 1898. Beiträge zur Kenntnis des Baues der Amphibiengliedmassen, besonders von Carpus und Tarsus. [Contributions to the knowledge of the amphibian limb, especially the carpus and tarsus.] Z wissenschatliche Zool 63: 61114.